One method. Evidence tested market by market.

Coverage tells us where filings exist. Local evidence tells us whether a ranked signal can become an actionable recommendation. They are not the same claim.

11 Aug 2026

Public sourceRegulator or exchange filing

NormalizeRole, amount, currency and dates

Score 0-100Six publication-time factors

Local rankCompared only with its market

Evidence gateWatch or actionable

Market

Period

Served strategy · top 10 · EU strict

Monthly top-10 backtest of the served ranking under strict point-in-time discipline, hold T+90, net of costs (audit 200).

Annual net return · max drawdown per year

2022+35.2%-26.6% max DD

2023+16.2%-17.0% max DD

2024-8.6%-19.3% max DD

2025+44.3%-12.2% max DD

2026 (Q1)+30.7%-1.4% max DD

Period summary · Since 202210,000 € → 27,069 €+26.4% CAGR · -26.6% max DD · 510 picks

Gold dot: 2025-2026 ran on an exceptional regime, never a guaranteed base rate.

Net of a 0.6% round-trip cost, returns winsorized at +-50%, T+90 hold. Figures were deliberately revised down in audit 200 to remove look-ahead bias: every input is point-in-time. Source: SCENARIO_RETURNS (audit 200). Backtest on public filings, not investment advice. Past performance does not predict future returns. Full methodology →

Test periods2023-2024 calibration · 2025 validation · 2026 Q1 shadow · T+90 net

A market needs two positive disjoint confidence bounds and sufficient price coverage before its signals become actionable recommendations. The explorer above publishes what can be measured today, cell by cell: below 30 observations it says n<30 instead of inventing a number.

See 43 covered markets and their public sources

FRAMF

USSEC (Form 4)

INSEBI (PIT + SAST)

KRDART (FSS)

DEBaFin

GBRNS (LSE)

CHSIX SER

CASEDI

AUASX

JPEDINET (Japan FSA)

ATFMA (OeKB OAM)

BALNasdaq Baltic

BEFSMA

BGBSE-Sofia (FSC)

BRCVM (Brazil)

CLCMF (Chile)

CNSSE / SZSE (CSRC)

CZCNB (Czech)

DKFinanstilsynet DK

EEFinantsinspektsioon (OAM)

ESCNMV

FIFIN-FSA

HKHKEX

HRHANFA / Zagreb SE

IDOJK / IDX

IEEuronext Dublin

ILTASE (ISA)

ITConsob

MYBursa Malaysia (SC)

NGNigerian Exchange (NGX)

NLAFM

NOOslo Børs

NZNZX

PHPSE

PKPSX (SECP)

PLKNF

ROASF / Bucharest SE

SATadawul (CMA)

SGSGX

THThai SEC (Form 59)

TRKAP / SPK (Borsa Istanbul)

TWTWSE / TPEx (MOPS)

ZAJSE

Σ · 0-100

How the ranking picks

0-100

The 6 signals

Zero parameters fitted on returns. The weights were frozen before the backtest.

1. Screen the universe

Companies above 300 M€ market cap, insider trades under 1% of float.

2. Score eligible filings

6 equal-weighted point-in-time signals, ranked cross-sectionally each month.

3. Hold the top names

Top 10 per month, equal weight, held 90 days (T+90).

Performance and transparency

The full report lives on one page: archived backtest, losing windows, statistical limits and the forward tracker.

Forward tracker: maturing, returns published from the threshold onwards.

See the full performance report

No one else covers the intersection.

OpenInsider covers the US for free. Quiver sells a US-only proprietary API. Bloomberg has it all, for $24k/yr. Sigma sits at the intersection: multi-market, open, backtested, AI-agent ready.

Capability	Sigma	OpenInsider	Quiver	Bloomberg
US Form 4 (SEC)	✓	✓	✓	✓
Non-US filings (42 other covered markets)	✓	✗	✗	✓ (paywall)
FX-normalised cross-market	✓	✗	✗	✓
Composite score (open formula)	✓	✗	proprietary	✗
Free interactive backtest viewer	✓	✗	paid	paid
Public REST API	✓ (free tier)	✗	paid	enterprise
MCP server (AI-native)	✓	✗	✗	✗
Entry price	€0 / €39	$0	$10+/mo	$24k/yr

Sigma vs OpenInsider Sigma vs Quiver Quant Pricing

Real examples

18 real captured signals, documented with their public filing and T+365 return (window Mar 2023 · May 2025).

See the case studies

Free account

Start reading the filings today

The free account opens the full read-only interface, no card required. Pro (€39/mo) is on an honest waitlist until payments open.

Create free account See pricing and the Pro waitlist

InsidersTradesSigma

Public insider buys,ranked into 10 names to watch.Every market shows its evidence, even when it is bad.

Point-in-time backtest51-month backtest (2022-2026Q1, point-in-time strict)

2022+35.2%-26.6% max drawdown

2023+16.2%-17.0% max drawdown

2024-8.6%-19.3% max drawdown

2025+44.3%-12.2% max drawdown

2026 (Q1)+30.7%-1.4% max drawdown

10,000 €27,069 €+26.4% average per year · -26.6% max drawdown

We publish the losing windows of our backtests, and we publish where the backtest and the live ranking differ.

Regime-dependent simulation on real data, not a live return. Point-in-time strict: the figures were deliberately revised down (audit 200) to remove look-ahead bias, and re-measured on the repaired corpus (audit 216). The live ranking has run this construction since 2026-08-16, but these figures remain a backtest: no live window exists yet, and the served book covers more venues than the backtest universe. Past performance does not predict future returns. How we measure →

We monitor public filings across 43 markets and score eligible purchases from 0 to 100. Each purchase is ranked within its own market.

Free today · Pro (€39/mo) on waitlistBacktest, not investment advice. Past performance does not predict future returns.

Official public sources

1,576,417 declarations

Updated daily

One method. Evidence tested market by market.

How the ranking picks

No one else covers the intersection.

Start reading the filings today

Public insider buys,ranked into 10 names to watch.Every market shows its evidence, even when it is bad.

One method. Evidence tested market by market.

How the ranking picks

Latest signals

28 markets shown out of 43 covered. One schema.

No one else covers the intersection.

Most active companies

Most active insiders

Latest transactions

Start reading the filings today

Public insider buys,ranked into 10 names to watch.Every market shows its evidence, even when it is bad.

Latest signals

28 markets shown out of 43 covered. One schema.

Most active companies

Most active insiders

Latest transactions